RESEARCH Open Access

Impact of coronavirus disease 2019 onpulmonary function in early convalescencephaseYiying Huang1,2†, Cuiyan Tan1†, Jian Wu1†, Meizhu Chen1, Zhenguo Wang1, Liyun Luo3, Xiaorong Zhou1,Xinran Liu1, Xiaoling Huang1, Shican Yuan1, Chaolin Chen1, Fen Gao1, Jin Huang1, Hong Shan2 and Jing Liu1,2*

Abstract

Objective: This study investigated the influence of Coronavirus Disease 2019 (COVID-19) on lung function in earlyconvalescence phase.

Methods: A retrospective study of COVID-19 patients at the Fifth Affiliated Hospital of Sun Yat-sen University wereconducted, with serial assessments including lung volumes (TLC), spirometry (FVC, FEV1), lung diffusing capacity forcarbon monoxide (DLCO),respiratory muscle strength, 6-min walking distance (6MWD) and high resolution CT beingcollected at 30 days after discharged.

Results: Fifty-seven patients completed the serial assessments. There were 40 non-severe cases and 17 severe cases.Thirty-one patients (54.3%) had abnormal CT findings. Abnormalities were detected in the pulmonary function tests in43 (75.4%) of the patients. Six (10.5%), 5(8.7%), 25(43.8%) 7(12.3%), and 30 (52.6%) patients had FVC, FEV1, FEV1/FVCratio, TLC, and DLCO values less than 80% of predicted values, respectively. 28 (49.1%) and 13 (22.8%) patients hadPImax and PEmax values less than 80% of the corresponding predicted values. Compared with non-severe cases,severe patients showed higher incidence of DLCO impairment (75.6%vs42.5%, p = 0.019), higher lung total severityscore (TSS) and R20, and significantly lower percentage of predicted TLC and 6MWD. No significant correlationbetween TSS and pulmonary function parameters was found during follow-up visit.

Conclusion: Impaired diffusing-capacity, lower respiratory muscle strength, and lung imaging abnormalities weredetected in more than half of the COVID-19 patients in early convalescence phase. Compared with non-severe cases,severe patients had a higher incidence of DLCO impairment and encountered more TLC decrease and 6MWD decline.

Keywords: COVID-19, Early convalescence, Lung function, Respiratory muscle strength

© The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License,which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you giveappropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate ifchanges were made. The images or other third party material in this article are included in the article's Creative Commonslicence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commonslicence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtainpermission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to thedata made available in this article, unless otherwise stated in a credit line to the data.

* Correspondence: liujing25@mail.sysu.edu.cn†Yiying Huang, Cuiyan Tan and Jian Wu contributed equally to this work.1Department of Pulmonary and Critical Care Medicine, The Fifth AffiliatedHospital of Sun Yat-sen University, 52 East Meihua Rd, Zhuhai City 519000,China2Guangdong Provincial Key Laboratory of Biomedical Imaging, The FifthAffiliated Hospital of Sun Yat-sen University, 52 East Meihua Rd, Zhuhai City519000, ChinaFull list of author information is available at the end of the article

Huang et al. Respiratory Research (2020) 21:163 https://doi.org/10.1186/s12931-020-01429-6

BackgroundCoronavirus Disease 2019 (COVID-19) is a new andhighly contagious respiratory disease caused by severeacute respiratory syndrome coronavirus 2 (SARS-CoV-2), which presented a risk of infection from human tohuman [1]. The current outbreak of COVID-19 hascaused a global pandemic. As of 7 June, 2020, there were6,663,304 confirmed cases and 392,802 confirmed deathsglobally. It might progress rapidly, and some patients de-veloped respiratory failure early in the disease. Theknowledge about COVID-19, including clinical manifes-tations, pathogenesis, even treatment came from re-search and observation during the acute infection period[2, 3]. In China, the vast majority of the patients hadbeen successfully discharged. Until now, no study havereported early prognosis in relation to the degree of lunginjury and rehabilitation in patients with COVID-19.Retrospective study showed that many patients had im-aging abnormalities when discharged, a few patientseven had pulmonary fibrosis. Lung function damage ofpatients with COVID-19 in early convalescence phasedeserves attention. In order to have a more comprehen-sive understanding of the possible clinical outcomes ofCOVID-19, we conducted a retrospective study involv-ing 57 discharged but undergoing rehabilitation COVID-19 patients. Serial lung function, lung imaging examin-ation and exercise capacity were examined at 30 daysafter discharged. In addition, we compared severe pa-tients with non-severe patients by outcome parameters.

Materials and methodsPatient selectionThis is a follow up study of COVID-19 patients at 30 daysafter discharged from our hospital. From January 17, 2020 toMarch 1, 2020, a total of 103 COVID-19 patients were ad-mitted to the Fifth Affiliated Hospital of Sun Yat-sen Univer-sity. The diagnosis of COVID-19 was based on the CDCcriteria. All patients had laboratory-confirmed SARS-CoV-2infection by real-time reverse transcription polymerase chainreaction (RT-PCR) or next-generation sequencing. They allreached uniform discharge standard issued by the NationalHealth Commission of China and had been released fromthe hospital over 1 month. In 30 days after discharged, pa-tients were eligible to participate in the study if they wereover 18 years of age. Patients with a previous history of pul-monary resection, neurological disease, or mental illness wereexcluded from our study. We obtained written informedconsent from the patients before pulmonary function testing.This study was approved by the institutional ethics commit-tee of the Fifth Affiliated Hospital of Sun Yat-sen University.

ClassificationWe retrospectively analyzed the medical records of thesepatients, and divided them into non-severe and severe

groups according to the severity of the disease. Patientswould be defined as severe cases if satisfied any of thefollowing criteria: shortness of breath, RR ≥ 30 times perminute; blood oxygen saturation ≤ 93% in resting state;partial arterial oxygen pressure (PaO2)/ fraction of in-spiration O2 (Fi02) ≤ 300mmmHg;respiratory failure re-quires mechanical ventilation; shock occurred orcombined with other organ failure required ICU moni-toring and treatment. Otherwise were mild cases.

Lung imaging acquisition and CT quantitative evaluationAll subjects underwent high resolution spiral CT(SOMATOM Definition Flash Siemens; Erlangen,Germany) scans in the supine position during end-inspiration. Images were reconstructed at 1.0 mm slicethickness, with 1 mm increment, 512mm × 512mm. Theimages were assessed by two radiologists, both of whomwere blinded to the clinical information. We used thesame method as Michael et al., to quantify pulmonaryinflammation severity [4–6]. Briefly, each of the five lunglobes was assessed for degree of involvement and classi-fied as none (0%), minimal (1–25%), mild (26–50%),moderate (51–75%), or severe (76–100%). No involve-ment corresponded to a lobe score of 0, minimal in-volvement to a lobe score of 1, mild involvement to alobe score of 2, moderate involvement to a lobe score of3, and severe involvement to a lobe score of 4.An overalllung “total severity score” was reached by summing thefive lobe scores (range of possible scores, 0–20). All CTscores were independently performed by two respiratorydoctors. Agreement was reached by consensus.

6min walk testSix min walk test (6MWT) is an exercise test that evalu-ates the functional status which is relevant to daily activ-ities of patients with cardiopulmonary disease. Thewalking distance is closely related to gender, age andheight, conventionally need a hierarchical analysis ac-cording to the above parameters. However, the samplesize of our study was small which was not suitable forstratified analysis based on age, gender and height. Sowe estimated the walking distance of healthy people ofthe same gender, age and height according to referenceequations for the 6MWT in healthy adults [7]. Then wecalculated the ratio of measured value of the patients tothe predicted value of the healthy person in fair condi-tion. By comparing the ratio of two groups we could seewhether there was difference in 6MWD between non-severe and severe COVID-19 patients.

Pulmonary function test and respiratory muscle strengthmeasurementEach subject underwent a standard pulmonary functiontest (Master Screen, Jaeger, German). Recorded

Huang et al. Respiratory Research (2020) 21:163 Page 2 of 10

parameters include: total lung volume (TLC), forced vitalcapacity (FVC), residual volume (RV), forced expiratoryvolume in the first second (FEV1), maximum expiratoryflow rate (MMEF75/25), FEV1 / FVC ratio, and diffusingcapacity of the lung for carbon monoxide (DLCO).Impuse oscillation system (IOS) was used to measureairway viscosity resistance at an oscillation frequency of5 Hz(R5), and central airway resistance at an oscillationfrequency of 20 Hz (R20). Mouth pressure gauges canmeasure the maximum static inspiratory pressure(PImax) or maximum static expiratory pressure (PEmax)through a flanged cigarette holder. All subjects used thissimple method to gauge inspiratory and expiratorymuscle strength. The spirometry, DLCO, and respiratorymuscle strength measurements were expressed as a per-centage of predicted normal values.To protect lung function laboratory staff, lung func-

tion tests were performed in a room with negative pres-sure device. Staff wore personal protective equipment,including N95 respirators, protective glasses, gloves andgowns. In addition, each patient used disposable virusand bacterial filters during the test.

StatisticsStatistical analysis was performed using Statistical Pack-age for Social Science (SPSS) Version 13.0. Measurementdata was expressed as mean ± standard deviation. Con-tinuous variables were compared using independent-

sample t test, whereas the rank sum test was used fornonparametric data. Comparison of proportion was eval-uated by Chi-square test. Spearman correlation test wasused to detect the correlations between lung functionand lung total severity score. All statistical tests weretwo tailed. Statistical significance was taken as p<0.05.

ResultsCharacteristics of the enrolled COVID-19 patientsThis study evaluated a total of 102 patients. Five patientswere excluded for underage. Twenty-four patients werenot included as it was less than 30 days after discharge.Three patients were excluded due to neurological ormental illness. In addition, eight patients had been outof contact. At last, 57 patients had been included andcompleted the serial assessments in the study (Fig. 1).There were 26 men and 31 women with a mean age of46.72 ± 13.78 years (age range, 19 to 71 years),the meanbody mass index was 23.99 ± 3.55 kg / m2.Among the 57subjects, 46 (80.7%) had a history of direct contact withWuhan, Hubei. Nine patients(15.7%)had a history ofsmoking. Twenty-one patients (36.8%) had preexistingmedical illness. The four most common preexisting ill-nesses were hypertension (11 patients), diabetes (fourpatients), malignant tumor (three patients) and cardio-vascular disease (three patients). All of these conditionswere either healed, or stable and well controlled at the

Fig. 1 Enrollment of COVID-19 patients in Early Convalescence

Huang et al. Respiratory Research (2020) 21:163 Page 3 of 10

time of testing during the study. No patient was reportedhaving chronic respiratory diseases.Among all subjects, seventeen were severe cases

(29.8%), forty were non-severe cases (70.2%). There weremainly male patients(70.6%)in the severe group, and theaverage age of patients was older compared with non-severe cases. The mean Pao2/Fio2 ratio among severecases was significantly lower than non-severe cases(198.47[SD, 97.04]; 355.51[SD, 37.23], P<0.001). Mean-while, severe cases had higher serum lactate

dehydrogenase (LDH), C-reactive protein (CRP) peaksand lower lymphocyte count compared with non-severecases. But there was no significant difference in thevalues of white blood cells, creatine kinase (CK), lacticacid peaks and length of hospitalization between the twogroups (Table 1).

Lung function tests and respiratory muscle strengthTable 2 presents the results of pulmonary function testsand respiratory muscle strength among COVID-19

Table 1 Demographic and clinical characteristic of COVID-19 patients

Characteristic Total (n=57) Severe (n=17) Non-severe (n=40) p Value

Age, year 46.72 (13.78) 52.53 (13.30) 44.25 (13.3) 0.031

Male gender, No 26 (45.6%) 12 (70.6%) 14 (35.0%) 0.014

BMI, kg/m2 23.99 (3.55) 25.54 (3.43) 23.33 (3.42) 0.103

Pre-existing medical illness 21 (36.8%) 7 (41.2%) 14 (35%) 0.658

LOS, days 20.89 (7.22) 20 (16-24) 19 (15-24) 0.834

WBC, ×10^9/L 5.01 (1.50) 4.47 (1.35) 5.24 (1.52) 0.076

lymphocyte count, ×10^9/L 1.60 (0.55) 1.30 (0.35) 1.72 (0.58) 0.008*

CRP, mg/dL 9.69 (13.77) 22.65 (18.19) 4.18 (5.66) <0.001*

LDH, U/L 175.47 (43.60) 201.94 (43.96) 164.22 (38.76) 0.002*

CK, U/L 91.95 (118.16) 133.18 (209.4) 74.42 (31.69) 0.235

Lactic acid, mmol/L 1.59 (0.61) 1.51 (0.65) 1.62 (0.59) 0.511

PaO2 to FiO2 ratio, mmHg 308.67 (94.40) 198.47 (97.04) 355.51 (37.23) <0.001*

TSS on the worst chest CT scan 4.28 (4.26) 8.59 (4.15) 2.45 (2.73) <0.001*

TSS on chest CT on the 14th day after discharge 1.75 (2.23) 3.94 (2.33) 0.83 (1.39) <0.001*

glucocorticoids use 16 (28.1%) 11 (64.7%) 5 (12.5%) <0.001*

Total methyprednisolone dosage, mg 213.75 (323.87) 289.09 (370.4) 48.0 (17.89) 0.019*

Values are expressed as mean (SD)*Statistically significant

Table 2 Results of pulmonary function tests and respiratory muscle strength among COVID-19 patients

Parameter Total (n=57) Severe (n=17) Non-severe (n=40) p Value

FVC (% of predicted) 100.96 (15.93) 95.92 (19.59) 103.10 (13.83) 0.12

FEV1 (% of predicted) 97.89 (14.91) 93.93 (16.79) 99.57 (13.92) 0.194

FEV1/FVC(%) 81.22 (6.13) 80.58 (4.88) 81.49 (6.62) 0.614

TLC (% of predicted) 93.94 (12.75) 88.72 (16.20) 96.22 (10.35) 0.048*

RV (% of predicted) 90.68 (28.08) 86.57 (23.96) 92.47 (29.82) 0.327

DLCO (% of predicted) 78.38 (13.59) 74.14 (18.85) 80.12 (10.56) 0.139

Raw(% of predicted) 105.38 (31.38) 99.46 (26.32) 108.03 (33.38) 0.524

R5(% of predicted) 126.64 (29.45) 118.75 (29.98) 130.00 (28.96) 0.072

R20(% of predicted) 132.76 (30.95) 120.15 (31.46) 138.12 (29.50) 0.024*

Pi max (% of predicted) 76.16 (24.28) 80.49 (29.24) 74.26 (21.93) 0.382

Pe max (% of predicted) 102.73 (32.68) 98.00 (27.11) 104.80 (34.96) 0.637

6MWD, m 561.97 (45.29) 517.43 (44.55) 573.52 (38.38) 0.012*

6MWD (% predicted) 94.61 (6.55) 88.46 (7.61) 96.20 (5.31) 0.011*

Values are shown as mean (SD) severe vs non-severe with p values*Statistically significant

Huang et al. Respiratory Research (2020) 21:163 Page 4 of 10

patients. During follow-up at 1 month after hospital dis-charge, there were 30 individuals (52.6%) with abnormaldiffusion capacity among the 57 patients participating inour study. According to the ATS recommendations forevaluating respiratory impairment [8], twenty-six pa-tients (86.7%) had mild impairment of DLCO, while theother four (13.3%) had moderate impairment. There wassignificant difference in impaired diffusing-capacity be-tween the two groups, which accounted for 42.5% innon-severe cases, and 75.6% in severe cases, respectively(p < 0.05, Table 3).The group means of forced expiratory volume in 1 s

(FEV1), static lung volumes were within normal limits(>80% predicted). However, several cases of abnormal-ities in FVC, FEV1, and FEV1/FVC ratio were detected.Five patients (8.7%) had mild impairment of FVC, one(1.8%) had moderate impairment of FVC, 5 (8.7%) hadmild impairment of FEV1, and 25 (43.9%) had mild im-pairment of FEV1/FVC. There were 8 patients (14.0%)and 10 patients (17.5%) had increased R5 and R20 morethan 150% of the predicted value, respectively. Up to12.2%(n = 7) of patients had reduction in parameters oflung volume (TLC) at 1 month. Among them, 6 hadmild impairment, one had moderate impairment. TLCdeclined more significantly in severe cases (p = 0.048).There was no difference in FVC, FEV1, and FEV1/FVCbetween the two groups. Table 4 shows the detailed pul-monary function data of all 57 subjects. The majority of

the impairment in FEV1 and FVC suggests a restrictiveabnormality. One patient without history of asthma hadobstructive abnormality with a FEV1/FVC ratio<70%predicted (up to 72% after bronchodilation), who hadsignificant history of cigarette smoking. Although nocomplained of symptoms of asthma, one other patienthad a significant bronchodilator response with incre-ments of FEV1>200 ml after inhalation of salbutamol.More than half of the subjects had impairment in re-

spiratory muscle strength. There were 28 patients(49.1%) and 13 patients (22.8%) had Pimax and Pemaxvalues less than 80% of the predicted value, respect-ively.13 patients had moderate impairment of respiratorymuscle strength, of whom 11 were non-severe cases(Table 4). When grouped by the administration of ster-oid, no statistical significance was found in respiratorymuscle strength between the glucocorticoid group andthe regular group (Table 5).

Chest radiographs and correlations with lung functionDuring follow-up at 30 days after discharge, six patients(10.5%) complained of slight cough, four (7.0%) hadshortness of breath, and three (5.3%) had occasionalwheezing. Follow-up CT scan at this time showed that31 patients (54.4%) had residual abnormality, of which16 were severe cases (94.1%) and 15 were non-severecases (37.5%).Most of the residual imaging abnormalitieswas patchy ground glass opacity with periphery

Table 3 The abnormal rate of pulmonary parameters and respiratory muscle strength between severe cases and mild cases

Characteristic Total (n=57) Severe (n=17) Non-severe (n=40) chisq p value

FEV1 < 80% of pred 5 (8.8) 3 (17.6) 2 (5.0) 1.066 0.302

FEV1 ≥ 80% of pred 52 (91.2) 14 (82.4) 38 (95.0)

FVC < 80% of pred 6 (10.5) 4 (23.5) 2 (5.0) 2.604 0.107

FVC ≥ 80% of pred 51 (89.5) 13 (76.5) 38 (95.0)

FEV1 / FVC < 80% 25 (43.9) 9 (52.9) 16 (40.0) 0.811 0.368

FEV1 / FVC ≥ 80% 32 (56.1) 8 (47.1) 24 (60.0)

TLC < 80% of pred 7 (12.3) 4 (23.5) 3 (7.5) 1.552 0.213

TLC ≥ 80% of pred 50 (87.7) 13 (76.5) 37 (92.5)

DLCO < 80% of pred 30 (52.6) 13 (76.5) 17 (42.5) 5.522 0.019*

DLCO ≥ 80% of pred 27 (43.4) 4 (23.5) 23 (57.5)

R5 ≥ 150 of pred 8 (14.0) 2 (11.8) 6 (15.0) 0.103 0.554

R5 < 150 of pred 49 (86.0) 15 (88.2) 34 (85.0)

R20 ≥ 150 of pred 10 (17.5) 3 (17.6) 7 (17.5) 0.000 0.631

R20 < 150 of pred 47 (82.4) 14 (82.4) 33 (82.5)

PImax < 80% of pred 28 (49.1) 9 (52.9) 21 (52.5) 0.001 0.976

PImax ≥ 80% of pred 29 (50.9) 8 (47.1) 19 (47.5)

PEmax < 80% of pred 13 (22.8) 4 (23.5) 9 (22.5) 0.007 0.592

PEmax ≥ 80% of pred 44 (77.2) 13 (76.5) 31 (77.5)

pred predicted*Statistically significant

Huang et al. Respiratory Research (2020) 21:163 Page 5 of 10

Table 4 Clinical and Pulmonary Function Data of COVID-19 Patients (n = 57)

Patient No Age, yr Clinical type FVC FEV 1 FEV 1 % FVC DLCO SB TLC R 5 R20

1 41 2 85.80 90.40 90.57 77.90 92.60 99.10 103.40

2 36 2 114.20 114.50 86.71 79.40 110.90 113.60 153.50

3 29 2 103.80 105.00 87.81 72.20 100.00 112.50 102.30

4 65 1 90.60 88.50 77.09 71.50 81.00 122.70 117.60

5 52 1 75.90 75.90 80.77 82.00 69.90 197.70 195.70

6 36 1 75.70 80.40 89.21 68.00 81.80 117.10 117.20

7 56 2 72.00 74.60 87.38 64.20 70.70 144.20 167.10

8 33 2 114.50 114.10 83.82 71.70 98.40 91.20 112.50

9 54 1 94.30 89.30 79.95 60.60 83.50 108.60 101.30

10 42 1 84.60 83.60 81.67 92.50 78.70 172.50 171.00

11 38 2 102.70 94.90 79.88 77.20 93.70 97.50 99.60

12 33 2 86.80 90.00 89.78 67.50 94.60 125.10 137.60

13 69 1 88.00 86.00 75.27 48.60 77.40 113.00 112.70

14 37 2 74.40 62.00 71.88 71.80 77.30 146.50 167.40

15 20 2 85.70 73.70 79.04 78.60 82.00 121.60 159.60

16 71 1 91.90 88.00 79.50 49.20 89.30 109.60 90.60

17 36 1 83.00 83.20 82.98 72.80 79.40 76.10 85.90

18 33 2 95.00 98.10 89.22 72.80 100.30 225.20 202.70

19 37 2 96.60 108.00 96.25 70.00 108.20 178.30 189.60

20 63 2 93.30 84.90 76.31 72.40 91.00 115.30 93.90

21 37 2 118.20 93.70 68.55 80.10 103.80 134.20 107.80

22 60 1 59.70 64.30 84.89 50.10 53.20 111.50 117.90

23 56 2 103.40 102.00 79.22 70.90 90.60 114.40 138.70

24 32 1 86.80 85.30 85.33 53.30 106.40 86.90 86.50

25 36 2 101.00 100.40 85.89 64.00 91.70 161.50 148.70

26 62 1 100.90 92.90 77.07 72.90 97.70 168.50 128.40

27 39 2 106.90 93.30 72.55 66.80 92.90 78.20 86.40

28 29 2 113.30 100.50 74.41 74.70 108.10 110.10 135.70

29 22 2 75.60 85.70 98.91 65.40 122.90 107.60 103.80

30 25 2 98.60 99.20 85.31 79.00 92.40 139.60 158.70

31 64 2 111.60 106.40 79.28 73.80 100.40 127.00 131.80

32 32 2 116.30 106.00 78.92 68.60 106.60 240.10 183.70

33 56 2 111.60 105.30 80.22 79.10 98.80 147.20 134.90

34 62 1 134.90 128.60 75.45 119.60 117.00 114.30 88.60

35 39 2 109.10 92.10 73.04 89.20 106.20 126.90 125.50

36 53 1 104.00 99.90 78.96 87.00 96.30 142.70 70.20

37 29 2 96.90 84.10 75.20 92.50 96.70 126.50 100.90

38 22 1 111.70 111.40 77.85 81.50 91.70 150.50 73.10

39 25 2 106.50 98.20 78.22 84.40 91.90 87.90 80.10

40 66 2 89.20 88.60 77.03 96.10 80.30 119.80 98.30

41 56 2 133.60 125.60 78.00 98.10 103.40 131.30 111.60

42 64 1 127.90 117.30 75.16 81.20 114.80 139.30 110.40

43 56 1 121.20 112.10 76.98 81.30 102.50 127.10 103.30

44 36 2 106.90 114.20 88.53 80.00 91.80 121.90 120.70

Huang et al. Respiratory Research (2020) 21:163 Page 6 of 10

distribution, which had obvious absorption comparedwith the worst chest CT scan (Fig. 2.a-b). Four patientshad pulmonary fibrosis (Fig. 2.c-d), all of whom were se-vere patients. Compared with non-severe cases, severepatients had a significantly higher CT score (3.94[SD,2.23]; 0.83 [SD, 1.39]; p < 0.01). At the acute phase, lungtotal severity score was negatively correlated with TLCand R20 (P = 0.049,0.044, Fig. 3), but the correlation dis-appeared during follow-up period.

6-min walk testThe mean 6-min walking distance (6MWD) in all sub-jects was 561.97 m (± 45.29 m). Severe patients had ashorter 6 min walking distance than non-severe patients(517.43 m [SD, 44.55 m]; 573.52 m [SD, 38.38 m], P =0.012). And the 6MWD of severe cases reached only88.4% of the predicted values, which was significantlower than non-severe cases (p = 0.011, Table 2).

DiscussionSince COVID-19 broke out worldwide over the last 6months, the mechanism, clinical characteristics, progno-sis and effective treatment of the disease had not yetbeen adequately elucidated despite the great efforts thathad been extended. Recent research and our dateshowed that nearly half of the discharged patients hadresidual abnormality in chest CT scan [6]. Global con-cerns are raised regarding the assessment of the lung in-jury for discharged patients. This study showed that inearly convalescence, approximately three-quarters of pa-tients with COVID-19 developed pulmonary functionimpairment, the most common of which was impaireddiffusing-capacity and the decline in FEV1/FVC ratio.DLCO abnormalities occurred in more than half of

the COVID-19 patients, the data indicated impaired

diffusion pathways in the intra-alveolar. To date, noother follow-up data on lung function in patients withCOVID-19 can be compared. S.A. MEO et al. re-ported that severe acute respiratory syndrome (SARS)and COVID-19 had similar biological and clinicalcharacteristics [9]. Previous studies on SARS survivorsshowed that impaired DLCO was the most commonabnormality, ranging from 15.5 to 43.6% [10–15]. Ourresults were consistent with them. Autopsy on pa-tients died from COVID-19 showed different degreesof destruction in alveolar structure, and pulmonaryinterstitial fibrosis were observed [16, 17].. Patho-logical changes in lungs can explain the impairedDLCO to a certain extent. Compared with non-severecase, severe patients were more likely to have DLCOabnormalities. Surprisingly, a small percentage of pa-tients with no residual imaging abnormalities also ex-perienced a slight decrease in DLCO. We think thatthese patients might have abnormal tiny blood vesselsor microthrombus formation. Long-term follow-upstudies of SARS survivors had shown that DLCOmight remain abnormal within 3 years of recovery insome patients [18]. We will continue to performlong-term follow-up on these patients to see thetrend of DLCO impairment.Our results showed that six patients (10.5%) had ob-

structive pulmonary dysfunction and 7 (12.3%) had re-strictive ventilation dysfunction. Two severe subjects hadresidual combined restrictive and obstructive type of func-tional impairment. Series articles on SARS survivorsreporting very low rates of either obstruction or restric-tion, which was consistent with our research data [10, 11].Pathological findings showed that mucous plugs werefound in small airway in some severe COVID-19 patients[17], which could explain the declined ventilatory function

Table 4 Clinical and Pulmonary Function Data of COVID-19 Patients (n = 57) (Continued)

Patient No Age, yr Clinical type FVC FEV 1 FEV 1 % FVC DLCO SB TLC R 5 R20

45 49 2 111.10 108.30 82.66 82.90 108.80 129.50 144.80

46 38 1 99.50 110.10 91.75 82.10 87.50 94.90 115.30

47 39 2 90.80 87.20 79.87 107.50 81.20 151.30 159.50

48 44 2 106.60 105.20 80.86 96.50 96.80 142.90 153.00

49 29 2 96.00 91.20 79.98 92.30 91.60 118.70 139.70

50 59 2 109.50 103.60 79.77 92.70 93.80 127.20 152.50

51 55 2 114.00 124.00 86.51 91.50 96.00 117.20 121.00

52 63 2 109.50 104.10 79.15 80.30 92.00 128.70 135.70

53 57 2 116.80 114.60 82.91 89.80 99.10 90.60 115.00

54 59 2 96.20 91.30 79.66 86.00 96.30 137.20 143.70

55 55 2 129.30 124.60 81.32 82.80 113.80 116.40 142.10

56 65 2 101.20 107.90 83.10 84.80 86.00 103.80 99.50

57 63 2 119.50 115.20 81.19 85.30 95.10 137.50 145.70

Values given as % predicted, unless otherwise indicated. Clinical type, 1 for severe, 2 for non-severe

Huang et al. Respiratory Research (2020) 21:163 Page 7 of 10

to an extent. In addition to acute lung injury, neuromus-cular weakness could also lead to decreased lung function.Certainly, a few patients suffered from lower FEV1 orFEV1/FVC ratio might due to long-term smoking or un-typical airway hyperresponsiveness.Surprisingly, during the early rehabilitation phase, lung

total severity score had no significant correlation withFEV1, FVC or DLCO (Fig. 3), which was inconsistentwith researches on SARS survivors [19]. It seems that

the impairment of lung function was not necessarilyagreed with the severity of illness or residual imagingchanges. It was an interesting finding. We speculate thatit was because most severe patients used glucocorticoidduring hospitalization, suggesting that corticosteroidsmay improve the prognosis of patients with COVID-19.But most of the subjects in our hospital were importedcases from Hubei, small sample size and selection biasmight affect statistical outcome. Besides, our CT

Table 5 Results of pulmonary function tests and respiratory muscle strength among COVID-19 patients between glucocorticoidgroup and the regular groups

Parameter All patients (n=57) GC group (n=16) Regular group (n=41) p Value

FVC (% of predicted) 100.96 (15.93) 97.25 (18.69) 102.40 (14.72) 0.414

FEV1 (% of predicted) 97.89 (14.91) 94.35 (15.40) 99.27 (14.68) 0.279

FEV1 / FVC (%) 81.22 (6.13) 80.74 (4.68) 81.40 (6.65) 0.804

TLC (% of predicted) 93.94 (12.75) 90.15 (16.01) 95.45 (11.06) 0.323

RV (% of predicted) 90.68 (28.08) 85.49 (19.01) 92.76 (30.95) 0.593

DLCO (% of predicted 78.38 (13.59) 74.67 (14.37) 79.78 (13.20) 0.657

Raw (% of predicted) 105.38 (31.38) 96.02 (25.81) 109.22 (32.93) 0.214

R5 (% of predicted) 126.64 (29.45) 119.66 (30.62) 129.37 (28.91) 0.127

R20 (% of predicted) 132.76 (30.95) 123.51 (31.99) 136.37 (30.15) 0.106

Pimax (% of predicted) 76.16 (24.28) 85.21 (26.54) 72.53 (22.65) 0.059

Pemax (% of predicted) 102.73 (32.68) 104.22 (28.03) 102.14 (34.68) 0.479

Values given as % predicted, unless otherwise indicatedGC glucocorticoid

Fig. 2 HRCT scan of a 44-year-old man in acute stage demonstrated bilateral peripheral ground-glass opacities (GGO). Lung total severity score(TSS)was 7. B. Follow-up CT of the same patient at 30 days after discharge from hospital showed that patchy ground glass opacity had obviousabsorption.TSS was 3 .C. Worst CT scan of a severe patient during acute stage showed diffuse GGO, consolidation also could be seen in somearea. TSS was 13. D: HRCT scan of the same patient obtained 30 days after discharge showed peripheral fibrosis consists of irregular linearopacities. Concomitant presence of GGO was also visible.TSS is 5

Huang et al. Respiratory Research (2020) 21:163 Page 8 of 10

quantitative evaluation was not actual represented thepercentage of lung parenchyma that showed evidence ofabnormalities. According to the criteria, in the samelung lobe, lung inflammatory lesions area within 25% ofthe difference was possible to be calculated as the samescore, which might influence statistical results. In thenext step, we will perform long-term follow-up studyand expand the sample size to see whether similar cor-relation conclusions still exist.More than half of the patients experienced a decrease in

respiratory muscle strength. Approximately 29.8% of pa-tients in our study were severe or critical, who had hypox-emia during hospitalization, requiring supplementaloxygen and bed rest, and prolonged bed rest might lead tomuscle disorders. In addition, systemic use of corticoste-roids might cause steroid myopathy. But when grouped bythe administration of steroid, no statistical significancewas found in respiratory muscle strength between theglucocorticoid group and the regular group. This result in-dicated that corticosteroid was not the main cause of re-spiratory muscle weakness. In fact, there was no differenceon declining respiratory muscle strength between severeand non-severe groups. However, the direct effect of viruson respiratory muscles needs further research.In early convalescence, the 6MWD of the severe pa-

tients was significantly shorter than that of the non-severe patients, indicating that the severe patients havepoor exercise tolerance. Besides the impaired TLC andworse DLCO in severe group, we should also pay atten-tion to cardiac function of the patients. Exercise cardio-pulmonary function should be performed in furtherstudies. Series articles on SARS survivors showed theimpaired lung function existed till 1 year [10, 11]. Lon-ger follow-up on COVID-19 patients should be made toobserve the characteristic and change tendency of lungfunction and exercise tolerance.There are several limitations to this study. Firstly, this

is a cross-sectional study with small sample size in strati-fied analysis, only provides a short follow-up. The het-erogeneity of our findings is not comprehensive.Secondly, only 57 of 102 COVID-19 patients (56%) inour hospital had completed the serial assessments, andthe results might not be representative of the entiregroup. Lastly, although full lung function tests and6MWT were conducted in our patients, we did not per-form cardiopulmonary exercise testing, as many patientscomplained of generalized muscle weakness on follow-up. Meanwhile, labor intensity of CPET might be toohigh for patients in early recovery period.In conclusion, impaired diffusing-capacity, respiratory

muscle strength decrease, and lung imaging abnormal-ities were detected in more than half of the COVID-19patients in early convalescence phase. Compared withnon-severe cases, severe patients had a higher incidence

Fig. 3 Correlation of total severity score on worst CT (from a1 to h1)and follow-up CT (from a2 to h2) with pulmonary function parameters.

Huang et al. Respiratory Research (2020) 21:163 Page 9 of 10

of DLCO impairment and encountered more TLC de-crease and 6MWD decline. Longer follow-up studies inCOVID-19 patients should be performed to investigatethe clinical outcome of recovered COVID-19 patients.

AbbreviationBMI: Body mass index; CK: Creatine kinase; CRP: C-reactive protein;WBC: White blood cell count; LDH: Lactate dehydrogenase; COVID-19: Coronavirus disease 2019; DLCO: Diffusing capacity of the lung forcarbon monoxide; FVC: Forced vital capacity; FEV1: Forced expiratory volumein 1 s; LOS: Length of hospital stay; PImax: Maximum static inspiratorypressures; PEmax: Maximum static expiratory pressures; Raw: Airway resistant;R5: Airway resistance at an oscillation frequency of 5 Hz; R20: Airwayresistance at an oscillation frequency of 20 Hz; TLC: Total lung capacity;TSS: Total severity score; 6MWD: 6 min walk distance; 6MWT: 6-min walk test

AcknowledgementsYiying Huang, Cuiyan Tan and Jian Wu contributed equally to this work.

Authors’ contributionsLJ designed the study, WJ performed the research, TCY collected the data,HYY analysed data and wrote the paper. All authors read and approved thefinal manuscript.

Availability of data and materialsAll data generated or analysed during this study are included in thispublished article.

Ethics approval and consent to participateThis study was approved by the institutional ethics committee of the FifthAffiliated Hospital of Sun Yat-sen University.

Consent for publicationConsent for publication is not applicable.

Competing interestsThe authors declare that they have no competing interests.

Author details1Department of Pulmonary and Critical Care Medicine, The Fifth AffiliatedHospital of Sun Yat-sen University, 52 East Meihua Rd, Zhuhai City 519000,China. 2Guangdong Provincial Key Laboratory of Biomedical Imaging, TheFifth Affiliated Hospital of Sun Yat-sen University, 52 East Meihua Rd, ZhuhaiCity 519000, China. 3Department of Cardiovascular Medicine, The FifthAffiliated Hospital of Sun Yat-sen University, Zhuhai, China.

Received: 30 April 2020 Accepted: 18 June 2020

References1. Chen N, Zhou M, Dong X, Qu J, Gong F, Han Y, Qiu Y, Wang J, Liu Y, Wei Y,

et al. Epidemiological and clinical characteristics of 99 cases of 2019 novelcoronavirus pneumonia in Wuhan, China: a descriptive study. Lancet. 2020;395(1)507-13.

2. Huang C, Wang Y, Li X, Ren L, Zhao J, Hu Y, Zhang L, Fan G, Xu J, Gu X,et al. Clinical features of patients infected with 2019 novel coronavirus inWuhan, China. Lancet. 2020;395(10223):497–506.

3. Ding Q, Lu P, Fan Y, Xia Y, Liu M. The clinical characteristics of pneumoniapatients coinfected with 2019 novel coronavirus and influenza virus inWuhan, China. J Med Virol. 2020.

4. Chung M, Bernheim A, Mei X, Zhang N, Huang M, Zeng X, Cui J, Xu W,Yang Y, Fayad ZA, et al. CT Imaging Features of 2019 Novel Coronavirus(2019-nCoV). Radiology. 2020;295(1):202-7.

5. James H, MT. Hypothesis: angiotensin-converting enzyme inhibitors andangiotensin receptor blockers may increase the risk of severe COVID-19. JTravel Med. 2020:1-2.

6. Li K, Fang Y, Li W, Pan C, Qin P, Zhong Y, Liu X, Huang M, Liao Y, Li S. CTimage visual quantitative evaluation and clinical classification of coronavirusdisease (COVID-19). Eur Radiol. 2020.

7. SHERRILL PLEA. Reference equations for the six-minute walk in healthyadults. Am J Respir Crit Care Med. 1998;158:1384–7.

8. Wang CJ, Ng CY, Brook RH. Response to COVID-19 in Taiwan: big dataanalytics, new technology, and proactive testing. JAMA. 2020;323(14):1339-40.

9. Meo SA, Alhowikan AM, Al-Khlaiwi T, Meo IM, Halepoto DM, Iqbal M,Usmani AM, Hajjar W, Ahmed N. Novel coronavirus 2019-nCoV: prevalence,biological and clinical characteristics comparison with SARS-CoV and MERS-CoV. Eur Rev Med Pharmaco. 2020;24(4):2012.

10. Hui DS, Wong KT, Ko FW, Tam LS, Chan DP, Woo J, Sung JJY. The 1-yearimpact of severe acute respiratory syndrome on pulmonary function,exercise capacity, and quality of life in a cohort of survivors. Chest. 2005;128(4):2247–61.

11. Ong K, Ng AW, Lee LS, Kaw G, Kwek S, Leow MK, Earnest A. 1-yearpulmonary function and health status in survivors of severe acuterespiratory syndrome. Chest. 2005;128(3):1393–400.

12. AWNL K-CO. Pulmonary function and exercise capacity in survivors ofsevere acute respiratory syndrome. Eur Respir J. 2004;24:436–42 2004.

13. Su MC, Hsieh YT, Wang YH, Lin AS, Chung YH, Lin MC. Exercise capacity andpulmonary function in hospital workers recovered from severe acuterespiratory syndrome. Respiration. 2007;74(5):511–6.

14. Xie L, Liu Y, Fan B, Xiao Y, Tian Q, Chen L, Zhao H, Chen W. Dynamicchanges of serum SARS-coronavirus IgG, pulmonary function andradiography in patients recovering from SARS after hospital discharge. RespRes. 2005;6(1):5.

15. Zheng Z, Chen R, Wu H, Liu X, He W, Xu Y, Chen S, Li Y, Zheng J, Zhong N.Changes in pulmonary function in severe acute respiratory syndromepatients during convalescent period. Zhongguo Wei Zhong Bing Ji Jiu YiXue. 2005;17(6):329.

16. Yao X, Li T, He Z, Ping Y, Liu H, Yu S, Mou H, Wang L, Zhang H, F u W, et al.A pathological report of three COVID-19 cases by minimally invasiveautopsies. medRxiv, https://doi.org/10.3760/cma.j.cn112151-20200312-00193.

17. Xu Z, Shi L, Wang Y, Zhang J, Huang L, Zhang C, Liu S, Zhao P, Liu H, Zhu L,et al. Pathological findings of COVID-19 associated with acute respiratorydistress syndrome. Lancet Respir Med. 2020;8(4):420–2.

18. NGAI JC, Ko FW, Ng SS, To K-W, Tong M, Hui DS. The long-term impact ofsevere acute respiratory syndrome on pulmonary function, exercise capacityand health status. Respirology. 2010;15(3):543–50.

19. Hsian-He Hsu M, Ching Tzao MP, Chin-Pyng Wu MP, Wei-Chou Chang M,Chen-Liang Tsai M, Ho-Jui Tung P, Cheng-Yu Chen M. Correlation of high-resolution CT, symptoms, and pulmonary function inpatients duringrecovery from severe acute respiratory syndrome, symptoms, andpulmonary function in. Chest. 2004;126:149–58.

Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.

Huang et al. Respiratory Research (2020) 21:163 Page 10 of 10